Experimental evaluation of achromatic phase shifters for mid-infrared starlight suppression.

Phase shifters are a key component of nulling interferometry, one of the potential routes to enabling the measurement of faint exoplanet spectra. Here, three different achromatic phase shifters are evaluated experimentally in the mid-infrared, where such nulling interferometers may someday operate. The methods evaluated include the use of dispersive glasses, a through-focus field inversion, and field reversals on reflection from antisymmetric flat-mirror periscopes. All three approaches yielded deep, broadband, mid-infrared nulls, but the deepest broadband nulls were obtained with the periscope architecture. In the periscope system, average null depths of 4x10(-5) were obtained with a 25% bandwidth, and 2x10(-5) with a 20% bandwidth, at a central wavelength of 9.5 mum. The best short term nulls at 20% bandwidth were approximately 9x10(-6), in line with error budget predictions and the limits of the current generation of hardware.

Design of a nonnull interferometer for aspheric wave fronts.

The presence of highly aspheric wave fronts in an interferometer leads to a need for system calibration, and this calibration requirement affects the design of the interferometer. Dynamic range, vignetting, and the ability to characterize components all must be considered during the design stages. The interferometer must be designed with respect to wave-front propagation as opposed to reference sphere aberrations. A nonnull interferometer for measurement of aspheric transmitted wave fronts has been developed, and the design process is described. Transmitted wave fronts for a conformal window and a progressive-addition bifocal lens have been measured to demonstrate the applicability of the system to aspheric testing.

Iterative reverse optimization procedure for calibration of aspheric wave-front measurements on a nonnull interferometer.

The accuracy of interferometric measurements made with a nonnull configuration is degraded by test-dependent aberrations. Calibration of the data can be done with reverse optimization methods. An iterative reverse optimization process that improves efficiency and eliminates sensitive merit function weighting issues is described. The process is shown to calibrate a nonnull interferometric measurement of a wave front with more than 200 waves of departure to an accuracy of 0.16 waves peak to valley and 0.02 waves rms.